BEHAVIOR OF FIRE FLIES 265 



see the numerous flashes given by the males in the other jar 

 but that they could not see those of the females. The screen 

 was then removed so as to expose the females and it was found 

 that the behavior of the males in jar (B) changed almost im- 

 mediately. They became much more active and gave every 

 appearance of being much excited; and in three minutes they 

 produced a total of twenty-eight flashes of light as against 

 three in fifteen minutes before the females were exposed. Such 

 a specific stimulation of the male by the glow of the female is 

 very evident to anyone making observations on these creatures 

 and cannot be mistaken. It w T as observed again and again 

 throughout the season under many different circumstances. A 

 specific effect of the glow 7 of the female is also attested by the 

 fact that I was unable to induce the mating responses in the 

 male by artificial imitation of the flashes o c light produced by 

 the females. Many attempts were made mainly with an elec- 

 tric joocket lamp but in only two or three instances was there 

 any response on the part of the males used in the experiments, 

 and in these it was not at all definite, although in practically 

 all cases they responded definitely to the glows of females. 



There are several factors that may possibly be involved in 

 this phenomenon of recognition. Let us enumerate them: 

 (i) The female is always at rest when she glow r s while the male 

 is ordinarily on the wing; (2) the form of the luminous area 

 differs in the two sexes; (3) there is always a certain time rela- 

 tion between the glow of the male and that of the female in 

 response to it; (4) the quality of light produced by the two 

 sexes may differ; (5) the intensity may differ; (6) the dura- 

 tion of the glow 7 may differ. 



Many observations and experiments were made in attempting 

 to ascertain which of these factors are of importance in the 

 process in question. I shall first describe the results of some of 

 these and then attempt to interpret them. 



(a) The mating responses are quite as precise when both 

 male and female are at rest as they are when the male is on 

 the wing, and the males are no more likely to turn toward 

 other males. It is therefore evident that the fact that the female 

 is at rest while the male is usually in motion when they glow 

 can be of no particular value in the phenomenon of recognition. 



(b) At different times, as previously stated, a piece of ruby 



